CFDEMSolverIB

Submitted by xiaobaishu67 on Thu, 10/01/2015 - 17:32

Dear all,
I am simulating interaction of particle and fluid using CFDEMSOlverIB, but it stops here:
Time = 0.002
Selected 391658 cells for refinement out of 430289.
Refined from 430289 to 3171895 cells.
Selected 0 split points out of a possible 391658.
Courant Number mean: 92329.5 max: 7.85258e+09
- evolve()
timeStepFraction() = 1
Starting up LIGGGHTS
Step Atoms KinEng rke Volume
Executing command: 'run 100 pre no post no'
Step Atoms KinEng rke Volume
40101 8844 1.328017 0.069717623 0.000132651
CFD Coupling established at step 40200
40201 8844 0.42986637 0.046681598 0.000132651
Loop time of 0.48413 on 4 procs for 100 steps with 8844 atoms
LIGGGHTS finished
=== cfdemCloudIB::getDEMdata() === particle rotation not considered in CFD
nr particles = 8844

Do anybody know what i miss? Thanks!!

Best wishes
xiaobaishu67

Forums:

legendglh | Tue, 03/15/2016 - 01:14

Hi xiaobaishu67,

Any update? I have the problem as well. Thanks for any discussion.

Cheers,
Linhan

alice's picture

alice | Wed, 03/16/2016 - 14:16

Hello,
the line "Courant Number mean: 92329.5 max: 7.85258e+09" indicates that your calculation crashed for numerical reasons.
The reasons for this can be manifold, e.g. choice of boundary conditions, time step size, ...
You could for example first run a pure CFD-calculation witout particles and see if that works out.
Best regards,
Alice

gelinhan | Fri, 03/18/2016 - 07:04

Hi Alice,

Thanks for your reply. I am now trying to use the cfdemIB solver to simulate bubble and particle collisions. I set the fluid field rho as 1000 kg/m3 and nu as 1e-06 which is a common condition for water. Then I inserted one bubble (density 1.204 kg/m3, 1mm) and one particle (density 1500 kg/m3, 0.25 mm ) by liggghts into a still water zone with gravity of 9.81 m/s2. The velocity of particle could be well predicted by the solver but bubble motion seems to be unphysical with excessively large velocity at the beginning. I guess the too small density of air bubble causes some numeric errors. Could you give me some hints about how to solve this problem or what the reason could be?

Thanks for your attention and time. Any advice would be appreciated.

P.S. Is there any tested cases of cfdemIB solver with LES model?

Best regards,
Linhan